democrática” en la región

straightness or tilt. Our visual field provides such a framework-"retinal orien-

' '

Figure 75a

1 ,8

IOO F O R M

tation," I called it earlier. When the children and chimpanzees cocked their heads, they eliminated the tilt of the figure in relation to their visual field. But there is also "environmental orientation." When a painting on the wall hangs crookedly, we see the tilt even though we may tilt our heads correspondingly, as long as we refer the picture to the framework of the walls. Within the narrower world of the painting itself, however, the verticals and horizontals of the frame determine the two basic axes. In Figure 76, taken from an investi-

Figure 76

gation of space perception by Hertha Kopfermann, the inner figure, under

the influence of the tilted frame, tends to look like a tilted square, although by itself or within a vertical or horizontal frame, it looks like an upright diamond. In Figure 77, which comes from the ornamentation of a tablecloth in a Picasso still life, the diamonds have a tendency to look parallel to one another although objectively they differ in orientation. Children often draw the chimney per­ pendicular to the inclined edge of the roof even though this adherence to the more specific framework puts the chimney in an oblique position. As a rule,

then, the spatial orientation of units in a picture is determined by a number of

Figure 77

F O R M IOI

different influences. If a face is turned sideways, the nose will be perceived as upright in relation to the face but as tilted relative to the entire picture. The artist must see to it not only that the desired effect prevails, but also that the strength of. the various local frames of reference is clearly proportioned; they must either compensate one another or be subordinated to one another hier­ archically. Otherwise the viewer will be confronted with a confusing cross­ fire. Note the disturbingly indeterminate orientation of the central line in Figure 78.

Figure 78

In addition to the coordinates of the retinal field and those of the visual

environment, a third framework of spatial orientation is provided kinesthetic­ ally, by the muscular sensations in the body and the organ of equilibrium in the inner ear. In whatever position our body or head or eyes may be, we sense the direction of gravitational pull. In daily life these kinesthetic sensa­ tions are usually in harmony with those derived from the visual framework of the environment. But when one looks up at a tall building, even the awareness of the tilted head may not be quite sufficient to compensate for the apparent

backward tilt of the facade; and when the same view appears on a movie

screen, the observer's upright posture together with the upright picture frame make the photographed world look tilted.

Experiments by Herman Witkin have shown that people vary markedly in how much their spatial orientation relies on the visual sense and how much on the kinesthetic. The more visually responsive persons, taking their cues from the outside world, were found to be more generally outer-directed, more dependent on standards of the environment, whereas the more kinesthetically

102 F O R M

responsive persons, listening to the signals from within their bodies, seemed to be more inner-directed, following their own judgment rather than the tenets of the world.

So far I have referred to examples of moderate tilt, which often leaves the structural skeleton essentially unaltered. A turn of ninety degrees tends to interfere with the character of visual shapes more drastically by causing the vertical and the horizontal to exchange places. When a violin or sculpted figure is seen lying on its side, the symmetry axis loses much of its compelling strength, and the shape points in a lateral direction like a boat or arrow. Even more radical is the change when the object is turned upside down. The two figures in Figure 79 are both triangular but their shapes differ. Version a rises

Figure79

from a stable basis to a sharp peak; in version b a broad top balances heavily and precariously on a pointed foot.

These are dynamic changes, due to the direction of gravitational pull. The effect is greatest in objects for which dynamic expression determines visual identity most strongly, notably the human face. In surrealist films, faces are sometimes shown upside-down. The effect is frightening: even though we know better, visual evidence ·insists that we are seeing a new kind of face, a monstrous variation dominated by the blind opening of the mouth, thrusting forward with the raised prow of the nose, and displaying at the base two

rolling eyes, cradled in baggy lids, which close upward.

To be able to recognize objects regardless of their spatial position is, of course, advantageous. Young children seem to handle picture books with little consideration of whether the illustrations are right-side-up or upside-down; and it used to be assumed that quite generally the spatial orientation of objects did not matter either to children or to primitive tribesmen. Recent experiments have indicated, however, that under certain conditions pictures projected on the wall are more easily recognized by the young child when they arc right­ side-up, and that this difference tends to become irrelevant when the child

F O R M 103

reaches school age. At this point we cannot be certain just how much the recognition of visual objects is influenced by the modifications of perceptual

appearance accompanying change of spatial orientation.

In any case, to observe the spatial orientation of objects in the physical world is one thing; to draw pictures of them is quite another. This is particu­ larly true for young children. In the physical world they observe buildings, trees, and cars rooted to the ground, and they would be surprised to see people or animals standing on their heads. The empty space of the drawing paper, however, imposes no such constraints, and in the beginning one spatial orien­ tation seems to be as good as any other, e.g., for the depicting of human figures. Spatial orientation is not yet differentiated. Only gradually does the "correct" upright position impose itself, for reasons yet to be explorec;L One of them must surely be that under normal conditions, the retinal projection obtained from the upright picture corresponds to the one received when the child looks at the physical model. Furthermore, it is true even for the simple pictures pro­ duced by children that the one-sidedness of the gravitational pull introduces the distinction between up and down, which enriches our visual world both physically and symbolically. When modern painters or sculptors create works that can be looked at validly in any spatial position, they pay for this freedom by settling for a relatively undifferentiated homogeneity.

Projections

In the examples of spatial orientation thus far discussed one might have expected no change of visual identity since geometric shape had remained unaltered. Instead we noted that under certain conditions a new orientation will bring to the fore a new structural skeleton, which gives the object a dif­ ferent character. Turning now to deviations that involve a modification of geometric shape we find that "non-rigid" change may or may not interfere with the identity of the pattern, depending on what it does to the structural skeleton.

Cut a fairly large rectangle of cardboard and observe its shadow cast by a candle or other small light source. Innumerable projections of the rectangle can be produced, some of them looking like the examples of Figure So. Figure 8oa, obtained by placing the rectangle exactly at right angles to the direction of the light source, resembles the object very closely. All other angles of projection lead to more or less drastic deviations of appearance. Figure Sob, though devoid of symmetry and right angles, is readily seen as an undistorted rectangle, tilted in space. Here again the principle of simplicity is at work. Whenever a three-dimensional version of a figure is sufficiently stabler and

104

F O R M

Figure 80

more symmetrical than the flat projection, the observer will tend to see the simpler shape, extended in depth. Figure Soc is much less likely to be seen as the projection of the rectangle which in fact it is. As a flat upright, it has a vertical symmetry of its own. It is a rather simply shaped regular trapezoid, and the tension created by the unequal angles is compensated within the plane. Its structural skeleton does not point to a rectangle.

Figure Sod, finally, is no longer a projection of the rectangle at all, but rather one of the thickness of the piece of cardboard. One can understand intellectually that this view, too, is derived from our object, but the deviation can no longer be

seen.

This problem, specific to the perception of three­ dimensional objects, will be taken up again shortly.

Looking at projections has confronted us with the phenomenon known as the

constancy of shape and size.

More often than not, perceptual constancy is interpreted by textbooks of psychology in a misleadingly simplified fashion. It is pointed out, correctly, that if we saw physical objects the way they are

projected on the retinas of the eyes, they would undergo dreadful amoebic

transformations of shape and size every time the objects changed their position toward us or we changed our position toward them. Fortunately this does not happen. The percept produced by the brain from the retinal projection is such that we see the object as it

is

physically. Asked what he sees when shown the shadow of our tilted cardboard rectangle, a person will tell us that he

sees

a rectangle of constant, stable shape. Asked to draw a picture of it, he may well draw a rectangle.

All this is true enough, but the impression often given is that this particu­ lar "correction" of the stimulus pattern occurs automatically and universally, although not quite completely, and that it is due either to an inborn mech­ anism, which requires no further explanation, or to accumulated experience, which corrects the faulty retinal input on the basis of better knowledge. Ex­ periments such as those by T. G. R. Bower have shown that infants between

F O R M

105

two and twenty weeks of age discriminate among test objects, e.g., cubes, according to their objective size, and see tilted rectangles as rectangles and not according to the shape of their retinal projection. This shows that at least the elements of shape and size constancy are already present at an early age. How­ ever, this is not really the main point of interest.

Another glance at Figure 80 reminds us that by no means all projections are perceived according to objective shape, and the same is true for size. All depends on the particular nature of the projection and the other conditions prevailing in the given situation. Depending on these conditions, there may be compelling constancy, or none at all, or some intermediate effect. No matter whether constancy processing is inherent in the nervous system or acquired by experience, there must be in either case an intricate mechanism equipped to deal with the input data appropriately. We need to know two things: (1) what kind of projection leads to what kind of percept, and (2) by what prin­ ciples operate the mechanisms that do the processing?

What matters to the artist in particular is to know which shapes will

produce which effect. He can acquire this knowledge by studying the prin­ ciples at work in shape perception. To be sure, the visual conditions prevailing in daily life are by no means identical to those prevailing in a drawing or painting. Instead of the isolated projections picked out in Figure 80, for ex­ ample, in the physical environment one more commonly experiences whole sequences of continuously changing projections, and this increases the con­ stancy effect considerably. When the cardboard square shifts gradually from position to position, momentary projections support and interpret one another. In this respect, immobile media such as drawing, painting, or photography

are quite different from the mobile ones. A projection that, frozen in its

momentary aspect, looks compelling, mysterious, absurd, or unrecognizable passes by unnoticed as a mere phase in a sequence of changes when an actor moves on the stage or in a film, or when the camera or a human observer moves around a piece of sculpture. In experiments on the shape and depth perception of infants, a most influential factor proved to be the motion parallax, i.e., the changes of spatial appearance caused by the movements of the viewer's head. Figure

8oa

indicated that as long as we deal with a flat object, such as a cardboard rectangle, there exists one projection that does such complete justice to the visual concept of the object that the two can be considered identical­ namely, the orthogonal projection, obtained when the plane of the object is hit by the line of sight at a right angle. Under this condition, the object and its retinal projection have roughly the same shape.

1o6 F O R M

things, because their shapes cannot be reproduced by any two-dimensional

In document CASO TIPO No. 1B. Por la Vida (página 32-39)